[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
"Polis, Chelsea B." <cpolis@jhsph.edu> |

To |
"statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |

Subject |
st: Sources focused on time-varying covariates in survival analysis? |

Date |
Fri, 8 May 2009 13:32:00 -0400 |

Many thanks to all who responded! One further question based on the readings recommended below: I do have the Cleves, Gould, and Gutierrez book mentioned. In fact, I practically sleep with it under my pillow - it's been an absolutely invaluable guide for my dissertation. At the same time, I find myself wishing for a reference that delves even more deeply into survival analyses using time-varying covariates. Some sources address time-varying covariates, but seem to stop just shy of the full explanations given for other circumstances. Such a resource would be useful - even on these boards I have noticed that things like the potentially extraneous nature of tvc() function (if your data has multiple records per subject) has confused several people. Perhaps it is my own thick-headedness, but I found this point difficult to understand both from the STATA help and from the Cleves book - it was looking through responses on this board from Roberto Gutierrez that set this piece straight for me. Simply trying to check the proportionality assumption in my analyses has left me in a state of semi-permanent mild panic that I am erroneously applying advice for different issues to my own analysis - which involves not only time-varying covariates, but also a time-varying exposure. Which is a long-winded way of asking --- does anybody know of any resources which provide an in-depth treatment of time-varying covariates in survival analysis using STATA? It would be much appreciated! And once again, I have to say - I don't know what I would do without the benefit of having this group available as a resource. I am extremely grateful for its existence, and for all of you gurus who take the time to help point us in the right direction! Best regards, Chelsea Polis -----Original Message----- From: Carlo Lazzaro [mailto:carlo.lazzaro@tiscalinet.it] Sent: Thursday, May 07, 2009 1:53 AM To: statalist@hsphsun2.harvard.edu Cc: Polis, Chelsea B. Subject: R: RE: Why don't my IRs and Cox HRs echo each other? Dear Chelsea, folowing Kieran's helpful advice, besides Kaplan-Meier estimates, you could take a look at whether deaths in one out of the compared groups tended to occur earlier or later vs the other one via log-rank test (more sensitive to differences occurring at the end of the follow-up) and Wilcoxon test (more sensitive to differences occurring at the beginning of the follow-up). Amongst other contributions, this topic is well covered in Maarten L. Buis. An introduction to Survival Analysis. 2006 (http://home.fsw.vu.nl/m.buis/) Cleves MA, Gould WG, Gutierrez R. An Introduction To Survival Analysis Using Stata. Revised edition. College Station: StataPress, 2006. HTH and Kind Regards, Carlo -----Messaggio originale----- Da: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Polis, Chelsea B. Inviato: giovedì 7 maggio 2009 1.35 A: statalist@hsphsun2.harvard.edu Oggetto: st: RE: Why don't my IRs and Cox HRs echo each other? I apologize for the formatting of my table, it looked ok when I sent it. Please let me try again. Variable Deaths PY at risk IR HR 95% CI p-value HC use 0.07 No 91 1262.7 7.21 1.00 Yes 13 293.0 4.44 0.58 0.32-1.04 Current age 0.38 15-24 20 394.0 5.08 1.00 25-34 49 711.8 6.88 0.73 0.43-1.24 35+ 35 449.9 7.78 0.68 0.38-1.20 Sex partners in past year 0.01 None 18 241.2 7.46 1.00 One 76 1204.6 6.31 1.31 0.78-2.21 Two+ 10 109.9 9.10 3.40 1.54-7.54 -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Polis, Chelsea B. Sent: Wednesday, May 06, 2009 7:29 PM To: statalist@hsphsun2.harvard.edu Subject: st: Why don't my IRs and Cox HRs echo each other? Dear statalisters, I am doing survival analysis on time to death with time-varying covariates on an open, population -based cohort study. The base sample is essentially a census of individuals in 56 villages, and I am utilizing information from all female incident HIV seroconverters. I computed incidence rates based on the raw data ((number of deaths/person-time at risk)*100 - I obtained time at risk using the stdes command), but the IRs don't seem to echo trends in the univariate Cox HRs. In the sample data below, things appear reasonable for HC use (deaths per 100 person years is lower if HC=yes, and the HR reflects this). But for current age, deaths are higher in the 25-34 category than in the 15-24 category, but the HR trends suggests that being 25-34 is protective (though not significantly). Also, the magnitude seems off, for example, in the variable "Sex partners in past year" - having two or more seems to more than triple the hazard in the Cox regression, but merely increases from 7.46 to 9.10 in the deaths per 100 p-y. Am I missing something in expecting these numbers to echo trends in each other? Is this just a matter of non-significance within individual categories? Or a difference in time-to-event versus person-time analysis? Or because I am doing an analysis with time-varying covariates? Should I not expect these to align? Any help is appreciated! Variable Deaths PY at risk Deaths per HR 95% CI p-value 100 p-y HC use 0.07 No 91 1262.7 7.21 1.00 Yes 13 293.0 4.44 0.58 0.32-1.04 Current age 0.38 15-24 20 394.0 5.08 1.00 25-34 49 711.8 6.88 0.73 0.43-1.24 35+ 35 449.9 7.78 0.68 0.38-1.20 Sex partners in past year 0.01 None 18 241.2 7.46 1.00 One 76 1204.6 6.31 1.31 0.78-2.21 Two+ 10 109.9 9.10 3.40 1.54-7.54 * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**st: R: Sources focused on time-varying covariates in survival analysis?***From:*"Carlo Lazzaro" <carlo.lazzaro@tiscalinet.it>

**References**:**st: RE: Why don't my IRs and Cox HRs echo each other?***From:*"Polis, Chelsea B." <cpolis@jhsph.edu>

**st: R: RE: Why don't my IRs and Cox HRs echo each other?***From:*"Carlo Lazzaro" <carlo.lazzaro@tiscalinet.it>

- Prev by Date:
**st: correlation using data from multiple imputation datasets** - Next by Date:
**st: R: Sources focused on time-varying covariates in survival analysis?** - Previous by thread:
**st: R: RE: Why don't my IRs and Cox HRs echo each other?** - Next by thread:
**st: R: Sources focused on time-varying covariates in survival analysis?** - Index(es):

© Copyright 1996–2016 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |